Mahesh P.

Key Skills

Software

No software listed

Top Subject Matter

Autonomous driving and mapping (HumanLoop2, NVIDIA)

Autonomous driving and mapping (DeepMap)

Top Data Types

3D Sensor

Image

Top Task Types

Tracking

Mapping

Freelancer Overview

Process Executive — Data annotation specialist for HumanLoop2 datasets. Brings 6+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include HumanLoop2 (NVIDIA) and DeepMap projects. Education includes Bachelor of Engineering, G H Raisoni College of Engineering and Higher Secondary Certificate, NSB College. AI-training focus includes data types such as 3D Sensor, Geospatial, and Tiled Imagery and labeling workflows including Tracking and Mapping.

Labeling Experience

Process Executive — Data annotation specialist for HumanLoop2 datasets

3D SensorTracking

Performed annotation of image, video, and 3D LiDAR data for autonomous driving datasets with a focus on object tracking and sequence labeling. Ensured lane-change, scene labeling, and top-view free space outputs followed strict SOPs and quality guidelines. Conducted pre-label issue correction including alignment, height, occlusion, and topology inconsistencies.• Labeled lane boundaries and navigable boundaries including splits, merges, and paths.• Executed HumanLoop2 dataset labeling tasks for NVIDIA autonomous driving workflows.• Applied validation checks to resolve errors and maintain zero-defect dataset quality.• Mentored new team members using quality reports and onboarding support.

2022 - 2026

Process Executive — MapGT creation, QA, and validation (DeepMap)

Mapping

Performed map-based ground truth (MapGT) creation and validation for autonomous driving and mapping use cases during DeepMap-related labeling activities. Verified lane geometry, navigable boundaries, intersections, and key road attributes to improve downstream model training quality. Corrected pre-labeled data issues and performed QA to ensure zero-defect labels for validation pipelines.• Created and validated MapGT for DeepMap workflows.• Verified lane geometry and navigable boundaries including intersection and road attribute consistency.• Corrected alignment, occlusion, height, and topology errors in existing labels.• Supported collaboration with engineering teams and contributed to quality reporting.

2021 - 2022

Education

Work History